ClawHub
Back to skill

Security audit

Yufluent Clawhub Publish Yufluentcn Comp Scrape

Security checks across malware telemetry and agentic risk

Overview

This appears to be a cloud-backed competitor-analysis skill whose network upload behavior matches its stated purpose, with privacy precautions users should understand.

Install only if you are comfortable sending the selected competitor/export data and listing text to Yufluent's cloud service. Keep API tokens in a trusted environment, watch for any TOKENAPI_BASE_URL override, and avoid using sensitive or confidential business data unless your organization permits that cloud processing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Tainted flow: 'url' from os.getenv (line 271, credential/environment) → requests.post (network output)

Critical
Category
Data Flow
Content
}
    if body is not None:
        kwargs["json"] = body
    return requests.post(url, **kwargs)


def _raise_for_status(resp: requests.Response) -> None:
Confidence
86% confidence
Finding
return requests.post(url, **kwargs)

Description-Behavior Mismatch

Medium
Confidence
79% confidence
Finding
The fallback path converts local payload into a generic /agent/turn orchestration request and injects a natural-language instruction to directly execute the named skill. That broadens the runtime behavior beyond a narrow comparison client: if the remote service interprets session fields or text more flexibly than expected, the code may trigger unintended agent capabilities or data handling paths, increasing attack surface.

Vague Triggers

Medium
Confidence
77% confidence
Finding
The trigger phrases include broad natural-language requests such as '这些竞品帮我分析' and 'competitor batch', which can overlap with ordinary conversation and cause the skill to activate unexpectedly. Because the skill collects files/data and sends content to a cloud harness, overbroad activation increases the chance of unintended data handling or exfiltration to the vendor service.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
This CLI reads user-supplied competitor CSV/export content and optional listing text, then sends that material to a remote cloud API via run_skill(). Although the metadata mentions cloud execution, the code itself provides no explicit runtime warning, confirmation, or data-minimization control before transmitting potentially sensitive business data off-host, creating a real privacy and data-governance risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.